public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug rtl-optimization/61673] New: [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390
@ 2014-07-02 14:53 jakub at gcc dot gnu.org
2014-07-02 15:19 ` [Bug rtl-optimization/61673] " jakub at gcc dot gnu.org
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2014-07-02 14:53 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61673
Bug ID: 61673
Summary: [4.9/4.10 Regression] Miscompilation of
_gnutls_hostname_compare on s390
Product: gcc
Version: 4.9.1
Status: UNCONFIRMED
Keywords: wrong-code
Severity: normal
Priority: P3
Component: rtl-optimization
Assignee: jakub at gcc dot gnu.org
Reporter: jakub at gcc dot gnu.org
Target: s390-linux
char e;
__attribute__((noinline, noclone)) void
foo (const char *x)
{
char d = x[0];
int c = d;
if ((c >= 0 && c <= 0x7f) == 0)
e = d;
}
int
main ()
{
const char c[] = { 0x54, 0x87 };
e = 0x21;
foo (c);
if (e != 0x21)
__builtin_abort ();
foo (c + 1);
if (e != 0x87)
__builtin_abort ();
return 0;
}
is miscompiled supposedly starting with r202393. Before combine we have:
(insn 7 6 8 2 (parallel [
(set (reg:SI 47 [ D.1404 ])
(and:SI (subreg:SI (reg/v:QI 44 [ d ]) 0)
(const_int 255 [0xff])))
(clobber (reg:CC 33 %cc))
]) gnutls2.c:8 455 {*andsi3_esa}
(expr_list:REG_UNUSED (reg:CC 33 %cc)
(nil)))
(insn 8 7 9 2 (set (reg:CCU 33 %cc)
(compare:CCU (reg:SI 47 [ D.1404 ])
(const_int 127 [0x7f]))) gnutls2.c:8 39 {*cmpsi_ccu}
(expr_list:REG_DEAD (reg:SI 47 [ D.1404 ])
(nil)))
(jump_insn 9 8 10 2 (set (pc)
(if_then_else (leu (reg:CCU 33 %cc)
(const_int 0 [0]))
(label_ref:SI 15)
(pc))) gnutls2.c:8 599 {*cjump_31}
(expr_list:REG_DEAD (reg:CCU 33 %cc)
(int_list:REG_BR_PROB 3900 (nil)))
-> 15)
and combine turns this into:
(note 7 6 8 2 NOTE_INSN_DELETED)
(insn 8 7 9 2 (parallel [
(set (reg:CCZ 33 %cc)
(compare:CCZ (and:SI (subreg:SI (reg/v:QI 44 [ d ]) 0)
(const_int -128 [0xffffffffffffff80]))
(const_int 0 [0])))
(clobber (scratch:SI))
]) gnutls2.c:8 453 {*andsi3_cconly}
(nil))
(jump_insn 9 8 10 2 (set (pc)
(if_then_else (eq (reg:CCZ 33 %cc)
(const_int 0 [0]))
(label_ref:SI 15)
(pc))) gnutls2.c:8 599 {*cjump_31}
(expr_list:REG_DEAD (reg:CCU 33 %cc)
(int_list:REG_BR_PROB 3900 (nil)))
-> 15)
The -128 should have been 128, we want to test the sign bit of QImode rather
than also any of the higher bits that contain garbage.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug rtl-optimization/61673] [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390
2014-07-02 14:53 [Bug rtl-optimization/61673] New: [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390 jakub at gcc dot gnu.org
@ 2014-07-02 15:19 ` jakub at gcc dot gnu.org
2014-07-07 11:11 ` rguenth at gcc dot gnu.org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2014-07-02 15:19 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61673
--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Created attachment 33050
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33050&action=edit
gcc49-pr61673.patch
Untested fix.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug rtl-optimization/61673] [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390
2014-07-02 14:53 [Bug rtl-optimization/61673] New: [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390 jakub at gcc dot gnu.org
2014-07-02 15:19 ` [Bug rtl-optimization/61673] " jakub at gcc dot gnu.org
@ 2014-07-07 11:11 ` rguenth at gcc dot gnu.org
2014-07-08 15:40 ` jakub at gcc dot gnu.org
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-07 11:11 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61673
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Priority|P3 |P2
Target Milestone|--- |4.9.1
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug rtl-optimization/61673] [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390
2014-07-02 14:53 [Bug rtl-optimization/61673] New: [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390 jakub at gcc dot gnu.org
2014-07-02 15:19 ` [Bug rtl-optimization/61673] " jakub at gcc dot gnu.org
2014-07-07 11:11 ` rguenth at gcc dot gnu.org
@ 2014-07-08 15:40 ` jakub at gcc dot gnu.org
2014-07-08 15:44 ` jakub at gcc dot gnu.org
2014-07-08 15:55 ` jakub at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2014-07-08 15:40 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61673
--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Tue Jul 8 15:39:36 2014
New Revision: 212364
URL: https://gcc.gnu.org/viewcvs?rev=212364&root=gcc&view=rev
Log:
PR rtl-optimization/61673
* combine.c (simplify_comparison): Test just mode's sign bit
in tmode rather than the sign bit and any bits above it.
* gcc.c-torture/execute/pr61673.c: New test.
Added:
trunk/gcc/testsuite/gcc.c-torture/execute/pr61673.c
Modified:
trunk/gcc/ChangeLog
trunk/gcc/combine.c
trunk/gcc/testsuite/ChangeLog
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug rtl-optimization/61673] [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390
2014-07-02 14:53 [Bug rtl-optimization/61673] New: [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390 jakub at gcc dot gnu.org
` (2 preceding siblings ...)
2014-07-08 15:40 ` jakub at gcc dot gnu.org
@ 2014-07-08 15:44 ` jakub at gcc dot gnu.org
2014-07-08 15:55 ` jakub at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2014-07-08 15:44 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61673
--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Tue Jul 8 15:43:39 2014
New Revision: 212365
URL: https://gcc.gnu.org/viewcvs?rev=212365&root=gcc&view=rev
Log:
PR rtl-optimization/61673
* combine.c (simplify_comparison): Test just mode's sign bit
in tmode rather than the sign bit and any bits above it.
* gcc.c-torture/execute/pr61673.c: New test.
Added:
branches/gcc-4_9-branch/gcc/testsuite/gcc.c-torture/execute/pr61673.c
Modified:
branches/gcc-4_9-branch/gcc/ChangeLog
branches/gcc-4_9-branch/gcc/combine.c
branches/gcc-4_9-branch/gcc/testsuite/ChangeLog
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug rtl-optimization/61673] [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390
2014-07-02 14:53 [Bug rtl-optimization/61673] New: [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390 jakub at gcc dot gnu.org
` (3 preceding siblings ...)
2014-07-08 15:44 ` jakub at gcc dot gnu.org
@ 2014-07-08 15:55 ` jakub at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2014-07-08 15:55 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61673
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution|--- |FIXED
--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2014-07-08 15:55 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-07-02 14:53 [Bug rtl-optimization/61673] New: [4.9/4.10 Regression] Miscompilation of _gnutls_hostname_compare on s390 jakub at gcc dot gnu.org
2014-07-02 15:19 ` [Bug rtl-optimization/61673] " jakub at gcc dot gnu.org
2014-07-07 11:11 ` rguenth at gcc dot gnu.org
2014-07-08 15:40 ` jakub at gcc dot gnu.org
2014-07-08 15:44 ` jakub at gcc dot gnu.org
2014-07-08 15:55 ` jakub at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).